Data on suspected risk factors to head and neck cancers will be collated from three distinct surveys and analyzed. The data comprise a rich source of information on smoking, drinking, dietary intake, medical symptomology and exposure to various environmental hazards for patients with a variety of cancer sites, as well as patients with a group of selected medical illnesses; similar data exist for a group of 3,000 suspected alcoholics. The data have been placed on magnetic tapes and are ready for analyses. Preliminary review of the quality of the data indicate that it is adequate. A number of hypotheses have been formulated for testing and are included under specific aims. Various statistical approaches will be used including covariance and discriminant analyses and standardization of rates. The inclusion of a wide variety of illnesses will permit a broad spectrum of research on the effects of alcohol.